A Robust Finite Element Simulator for Black-Oil and Fully Compositional Compressible Flow in Hydrocarbon Reservoirs

Abstract

Abstract We present a unified numerical simulator for reservoir engineering problems of the highest complexity: fully compositional and black-oil, three-phase, compressible flow on three-dimensional unstructured grid. In the context of black-oil modeling, where hydrocarbon components are lumped into a gas and an oil component with only the gas component being allowed to transfer between oil and gas phases, study cases may involve treating a variable bubble point pressure throughout the reservoir (e.g., in water flooding into saturated reservoirs, or introducing gas into an undersaturated grid cell). Traditionally, a primary variable switching strategy for phase appearance or disappearance has been proposed, which inherits convergence and phase identification issues. In this study, we adopt an overall molar composition-based framework that can self-consistently take care of such complications. We compute the black-oil properties from tables initially constructed through fully compositional phase split calculations. Phase properties across a broad range of pressures for different black-oil compositions are interpolated to correctly model the transitions between saturated and undersaturated states. While maintaining the accuracy, we are able to gain considerable speed-up in black-oil model -more pronounced in saturated reservoirs- as compared to the fully compositional model. In addition, previous black-oil models are mostly limited to low order numerical schemes. Here finite element (FE) methods are adopted, which are well suited to parallelization and provide flexibility in discretizing the geometry. Mass transport is updated explicitly by a locally mass conserving discontinuous Galerkin method. Globally continuous pressure and velocity fields are obtained through an implicit mixed hybrid FE scheme. The robustness and accuracy of our FE simulator are demonstrated in several complex problems, where we have attained considerable speed-up and maintained the accuracy by means of the new black-oil model, specifically for phase split computations.